2246
Solvent Effects upon the Phosphorescence Lifetimes and Photoreactivity of Butyrophenone R. David Rauh and Peter A. Leermakers’ Contribution from the Department of Chemistry, Wesleyan University, Middletown, Connecticut 06457. Received October 5, 1967 Abstract: Excitation of butyrophenone in moderately hydrogen bonding, polar glasses gives rise to dual component phosphorescence emission, one component relatively short lived (-0.001 sec) and other relatively long lived (-0.05 sec). On the basis of previous investigations, it is probable that these arise from the pure 3(n,n*) state in the former and from a state having both 3(n,n*) and 3(n,n*)characteristics in the latter. The long-lived component emission is undetected in hydrocarbon solvents, while in highly polar solvents the short-lived component is absent. The effect is best explained on the basis of local solvent effects. The photoreactivity of butyrophenone is also found to be a function of solvent polarity and is maximized in solvents where polarity is moder* in the reacting 3(n,a*) exately high, but not high enough to significantly introduce relatively inert n , ~character cited state.
T
he effect of solvent upon the excited states of aromatic carbonyl compounds has been a subject of current interest. 2--6 Recently, Lamo1a4 has demonstrated that the relative disposition of the 3(n,r*)and 3 ( 7 r , ~ * ) states of acetophenone is reversed in highly polar media compared to that in nonpolar media. The long-lived 3(7r,7r*) state becomes the triplet of lowest energy in 50% ethylene glycol-water (EGW), in a methylcyclohexane-silica gel slurry, in glacial acetic acid, in formamide, and in 85 % phosphoric acid. This is attributed to the rise in the 3(n,r*) and decrease in the 3(r,7r*) energy with increasing solvent polarity. Lamola also reported two types of phosphorescent emission from acetophenone in ethanol, a short-lived and a long-lived component, and suggested the possibility of two kinds of solvation sites.4 Yang has observed two phosphorescent states in 1-inadanone in both methylcyclohexane (MCH) and EPA (etherisopentane-ethanol), one of pure n , r * character and a longer-lived species having a mixture of n,r* and ~ , r * characteristics. Wagner * has to some extent correlated solvent polarity with the propensity of butyrophenone, valerophenone, and 2-hexanone to undergo the Norrish type I1 elimination. In all three cases a general increase in the quantum yield for disappearance of ketone in passing from hydrocarbon solvents to t-butyl alcohol was observed. The quantum yield based upon the formation of photoelimination products also shows a marked increase with solvent polarity, but only in media which are poor donors of radical hydrogen so that competing photoreduction is minimized.2 Much has been published on the effects of electrondonating substituents upon the lifetimes and photoThe reactive excited reactivity of aryl (1) Alfred P. Sloan Fellow, 1967-1969. (2) P. J. Wagner, Tetrahedron Leffers, 1753 (1967). ( 3 ) N. C. Yang and S. Murov, J . Chem. Phys., 45, 4358 (1966). (4) A . A. Larnola, ibid., 47, 4810(1967). (5) N. C. Yang, D. S. McClure, S. L. Murov, J. J. Houser, and R. Dusenbery, J . Am. Chem. SOC.,89, 5466 (1967). (6) J. N. Pitts, Jr.. H. W. Johnson, Jr., and T. Kuwana, J . Phys. Chem., 66, 2456 (1962). (7) J. G. Calvert and J. N. Pitts, Jr., “Photochemistry,” John Wiley and Sons, Inc., New York, N. Y.,1966, pp 382-389. (8) J . N. Pitts. Jr.. L. D. Hess. E. J. Baum. E. A. Shuck. J. K. S. Wan. P. A. Leerrnakers, and G. Vesley, Photochem. Phofobiol.,‘4, 305 (1965): (9) W. M. Moore, G. S.Harnmond, and R. P. Foss, J . Am. Chem. Soc., 83, 2789 (1961).
state in the photoreduction and the photoelimination of these compounds has been presumed to be the n , r * triplet. The relative energetic disposition of the n , r * triplet to the T,T* triplet in phenyl alkyl ketones becomes reversed upon the introduction of methyl, methoxy, and hydroxyl substituents causing the excited molecules to be less Furthermore, quantum yields based upon elimination products in substituted butyrophenone~~ and upon reduction products in substituted acetophenones5 have an inverse relationship to their phosphorescence lifetimes, lending additional evidence for the r,r* character of the lowest triplet state in the inert compounds. The pertinent data are given in Table I. Table I. Quantum Yields and Phosphorescence Lifetimes of Some Substituted Butyrophenones and Acetophenones
Substituent
None p-Me P-F p-OMe m-OMe p-OH 0-OH p-NHz
-Acetophenonea-. T ~ sec , $ketonec 0.004 0.084 .,. 0.26 0.71
...
...
...
0.68 0.66
... 0.04 0.006 ... ...
...
-Butyrophenoneb+,, sec $etiiylene 0.002 0.009 0.004 0.051
0.42 0.39 0.29 0.10
...
...
0.084
0.00 0.00 0.00
...
0.084
~
a All values for T~ and $ketone are reproduced from Yang’s recent paper.6 Photolyses were carried out in n-propyl alcohol. All values for r Pand for $eth, lene reproduced from Calvert and Pitts,7 p 383. The quantum yields measure the extent of photoelimination. Lifetimes were recorded in an ethanol glass, quantum yields in benzene. c At 3130 A.
In light of these investigations, the purpose of the present study is to demonstrate that solvent effects parallel the effects of inductive group substitution upon the photoreactivity of aryl ketones and to elucidate further the mechanism of solvent intervention in these reactions. Butyrophenone is an excellent compound for such a study. Not only do the 3(n,a*) and d(r,7r*) states lie very close in energy, as in acetophenone, but also the compound undergoes the Norrish type I1 elimination as well as photoreduction. The efficiency
Journal of the American Chemical Society / 90:9 / April 24, 1968
(10) G. S. Hammond and P. A. Leermakers, ibid., 84, 207 (1962). (11) G.Porter and P. Suppan, Pure Appl. Chem., 9, 499 (1964).
2247
of both reactions at room temperature may be correlated with the l/e phosphorescent lifetimes of butyrophenone in a variety of solvents at 77°K. One must be cautious in correlations of ambient temperature photochemistry with data about electronic states determined at much lower (liquid nitrogen) temperatures. In the research to be discussed, however, such correlations are not only meaningful but also very useful. We make no attempt to quantitatiuely correlate natural phosphorescence lifetime with photoreactivity. Our objective is to characterize excited states by their lifetimes-“long,” “short,” or both-at 77°K. On the assumption that solute-solvent interactions are not wholly different at 298 us. 77”K, we then infer that the emitting states (at 77’K) are essentially the same as the chemically active states at room temperature (at which no emission can be detected). We do want to introduce the qualification that many more solute-solvent contacts will be explored per unit time at room temperature so that interconversion of nearly degenerate triplet states will be more facile. Experimental Section Materials. Butyrophenone was obtained from Aldrich Chemical Company and was distilled at reduced pressure before use. All solvents were of spectroscopic or reagent grade and were used without further purification. Silica gel was Grace Davidson Commerical, 100-200 mesh, and was oven dried at 200”. Emission Lifetimes. Phosphorescence lifetimes were measured at 77°K on an Aminco-Bowman spectrophotofluorimeter with a rotating can attachment. The instrunlent was equipped with a Tetronix Type 561A oscilloscope with a plug-in Type 3A75 amplifier and a Type 2B67 time base. Decay curves were recorded using a Polaroid camera attachment. Photolysis. Photolyses- were carried out at room temperature with predominantly 2537-A light. A Rayonet reactor manufactured by Southern New England Ultraviolet Co. of Middletown, Conn., equipped with a merry-go-round apparatus was used. Cells were 8-mm diameter Vycor glass tubes sealed at one end and capped with serum caps through which a hypodermic needle outlet was placed to relieve internal pressure. Samples of 4 ml were used, each charged with 0.01 ml of dodecane which served as an internal standard. All reaction times were 90 min. Analysis was conducted using glpc techniques (column: 5 ft X 0.25 in., 10% silicon polar oil on Chromosorb support; column temperature, 150”; injector block temperature, 220”), and relative peak areas were estimated using the “cut out and weigh” method. Actinometers. The quantum yields based upon disappearance of butyrophenone and upon appearance of acetophenone in methanol are reported by Wagner as being 0.80 and 0.35, respectively, foroa 0.2 M degassed solution of butyrophenone irradiated at 3130 A. These results were checked using a benzophenone-benzhydrol actinometer.12 The quantum yields for the same solution were found to agree quite closely when the reaction was carried out under nondegassed conditions using a 2537-A source, being 0.78 and 0.33 for disappearance and appearance, respectively. Once this was determined, all samples were run under the same conditions and quantum yields were computed relative to these values.
Results and Discussion Table 11 provides a list of measured phosphorescence lifetimes for butyrophenone in a number of solvents which form a glass at 77’K. (Since quantum yields for emission were not determined, these can only be estimates of natural radiative lifetimes.) These solvents, or solvent systems, may be roughly divided into three categories: (1) those in which only a short-lived phosphorescent component is evident, l 3 (2) those which ex(12) W. M. Moore and M. Ketchum, J . Am. Chern. Soc., 84, 1368 (1962).
Rauh, Leermakers
Table II. Phosphorescence Lifetime. of Butyrophenone in Various Solvents
Solvent MCH 3-Methylpentane EPA 2-BuOH EtOH MeOH EtOH-EG 2:l 1:l 2:3 1 :5 EG W 9.5 : 1 4: 1 2.3:l 1:l 1:1.5 E t 0 H-HCI 0.5 Z 1.0% 5.0% MCH,Si02 EG-EtOH-HOAC, 1 :1 :2 HOAC-EGW 1 :1 (85 %)
rY (short component), sec
TP (long component), sec
x
10-3
x x x x
10-3 10-3 10-3
None None 2 . 0 x 10-2 1.5 X 3.0 x 3 . 5 x 10-2
0.8 1.0 3.0 3.0 3.5 7.0 8.0 8.0 8.0 7.0
x 10-3
x x x x
10-3 10-3 10-3
10-3
... ... ... None None
4 5
x lo-’ ... x 10-8
None 2.5 X 2.5 X None
3.0 3.0 3.0 3.5
x x x x
10-2 10-2
3.0 5.0 6.0 8.0 1.0
x x x x x
10-2
10-2
10-2 10-2 10-1
4 . 0 X 10-2 5 . 0 x 10-2 7 . 0 x 10-2 7.5 x 10-1 5.0 x 10-9 9 . 0 x 10-9 8 . 0 x 10-1
0 Lifetimes of excited states are uncorrected for nonradiative processes and thus are only estimates, since absolute and relative quantum yields for emission were not determined.
hibit two components, one short lived and one long lived, and (3) those which exhibit only a long-lived component. (Where two components were found, we were not able t o determine quantitatively the relative contribution of each. For our purposes this is important but not critical.) It is evident from Table I1 that there is no specific point at which the 3(n,7r*) and 3(7r,7r*) states are definitely reversed. In MCH and in 3methylpentane, only a short-lived triplet (-0.001 sec), presumably n,r* in character, is noticed. These solvents are not at all polar, thus the n electrons have little interaction with solvent in this case, so that it is not unreasonable that the lowest and emitting triplet is clearly n,ir* (see Figure la). In media of intermediate polarity, such as ethanol, methanol, t-butyl alcohol, and glacial acetic acid, solvation may surely take place in a variety of ways (as suggested by Lamola4 in the case of acetophenone in ethanol). Some molecules will then be in locally polar environments, others will not. The result will be equilibration of two possible nearly degenerate triplet species. Dual component emission is observed in these cases. The shorter lived component is pure K* -+ n in character and arises from molecules in mostly nonpolar solvent situations in which the carbonyl is left relatively unperturbed. The other emission arises from molecules whose n electrons are hydrogen bonded to the solvent, in which case the energy of the 3 ( n , ~ *state ) becomes nearly equivalent to that of the 3 ( ~ , 7 r * ) state. The energy of the latter state is correspondingly lowered due to solvent perturbation which leads to increased mixing (13) Yangs reports finding two phosphorescent components for 1indanone in methylcyclohexane. We find no ‘‘long-lived’’ component for butyrophenone in hydrocarbon solvents, but one may exist which is beyond our range of detection.
Solvent Eflects upon Phosphorescence and Phoforeactivify of Butyrophenone
2248 Table 111. Phosphorescence Lifetimes and Quantum Yields for Disappearance of Butyrophenone and Appearance of Acetophenone in Various Solvents
-
6 Solvent
Component 0.8 x 10-3 Est 0.001 3 x 10-3, 1.5
MCH n-Hexane r-BuOHb CHiCN HOAc EG-EtOH-HOAC, 1:1:2 EGW-HOAC, 1 :1 MeOH EG-EtOH, 5 :1 EtOH EtOH-0.5 HCI EtOH-1 % HCI EtOH-5 % HCI MCH-Si02 EGW Hap04 (85 %)
3.5 2.5
x x
...
T ~ sec ,
x
Butyrophenone 0.33 0.39@ 1.ma 0.9@ 1.0 1 .o
10-2
10-3, 3.0 x 10-2 10-3, 3 x 10-2
2.5 X 9.0 X 4 x 10-3, 3.5 x 10-2 2.5 x 10-3, 3.5 x 10-2 3.5 x 10-3, 3.0 x 10-2 4 x 10-3,4 x 10-2
...
5 x 10-3,8 8 x 10-2 1.0 x IO-' 8.0 x 10-1
x
0.94 0.78 0 74 0.66 0.48 0 46 0.33
10-2
-Aacetophenone/
Acetophenone
...
